Prognostic significance of c-Met expression in glioblastomas

Authors

  • Doo-Sik Kong MD, PhD,

    1. Department of Neurosurgery, Samsung Medical Center and Samsung Biomedical Research Institute, Sungkyunkwan University School of Medicine, Seoul, Korea
    Search for more papers by this author
    • The first 2 authors contributed equally to this article.

  • Sang-Yong Song MD, PhD,

    1. Department of Pathology, Samsung Medical Center and Samsung Biomedical Research Institute, Sungkyunkwan University School of Medicine, Seoul, Korea
    Search for more papers by this author
    • The first 2 authors contributed equally to this article.

  • Duk-Hwan Kim MD,

    1. Department of Molecular Cell Biology, Samsung Medical Center and Samsung Biomedical Research Institute, Sungkyunkwan University School of Medicine, Seoul, Korea
    Search for more papers by this author
  • Kyeung Min Joo MD, PhD,

    1. Department of Neurosurgery, Samsung Medical Center and Samsung Biomedical Research Institute, Sungkyunkwan University School of Medicine, Seoul, Korea
    Search for more papers by this author
  • Jin-San Yoo PhD,

    1. The Therapeutic Antibody Center, Korea Research Institute of Bioscience and Biotechnology, Daejeon, Korea
    Search for more papers by this author
  • Jong Sung Koh PhD,

    1. Center for Anti-Cancer Research, Korea Research Institute of Chemical Technology, Daejeon, Korea
    Search for more papers by this author
  • Seung Myung Dong PhD,

    1. Research Institute, National Cancer Center, Goyang, Gyeonggi, Korea
    Search for more papers by this author
  • Yeon-Lim Suh MD, PhD,

    1. Department of Pathology, Samsung Medical Center and Samsung Biomedical Research Institute, Sungkyunkwan University School of Medicine, Seoul, Korea
    Search for more papers by this author
  • Jung-Il Lee MD, PhD,

    1. Department of Neurosurgery, Samsung Medical Center and Samsung Biomedical Research Institute, Sungkyunkwan University School of Medicine, Seoul, Korea
    Search for more papers by this author
  • Kwan Park MD, PhD,

    1. Department of Neurosurgery, Samsung Medical Center and Samsung Biomedical Research Institute, Sungkyunkwan University School of Medicine, Seoul, Korea
    Search for more papers by this author
  • Jong Hyun Kim MD, PhD,

    1. Department of Neurosurgery, Samsung Medical Center and Samsung Biomedical Research Institute, Sungkyunkwan University School of Medicine, Seoul, Korea
    Search for more papers by this author
  • Do-Hyun Nam MD, PhD

    Corresponding author
    1. Department of Neurosurgery, Samsung Medical Center and Samsung Biomedical Research Institute, Sungkyunkwan University School of Medicine, Seoul, Korea
    • Department of Neurosurgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, 50 Ilwon-dong, Kangnam-gu, Seoul 135-710, South Korea===

    Search for more papers by this author
    • Fax: (011) 822-3410-0048


Abstract

BACKGROUD:

The authors investigated whether expression of c-Met protein in glioblastomas is associated with overall survival and biologic features representing tumor invasiveness in patients with glioblastomas.

METHODS:

Paraffin-embedded specimens of glioblastomas from 62 patients treated in a single institution were assessed by immunohistochemical (IHC) analysis of c-Met expression. On the basis of the clinical data for these patients, the association between c-Met expression and clinicobiologic features representing tumor invasiveness was analyzed.

RESULTS:

c-Met overexpression was detected in 29.0% (18 of 62) of glioblastomas. In patients with c-Met overexpression, the median survival was 11.7 months (95% confidence interval [95% CI], 9.9 months-13.5 months), compared with a median survival of 14.3 months (95% CI, 7.6 months-21.0 months) for patients whose tumors had no or little expression of c-Met (P = .031). On the radiographic analysis, 9 of 18 patients (50%) with tumors overexpressing c-Met demonstrated invasive and multifocal lesions on the initial magnetic resonance images, whereas only 9 of 44 patients (20.5%) with tumors that expressed no or little c-Met demonstrated these features (P = .030). Using immunohistochemistry, we also found a significant association between c-Met expression and matrix metalloproteinase-2,-9 (P = .020 and P = .013). Furthermore, Myc overexpression was found to be closely correlated with c-Met overexpression on IHC analysis (P = .004).

CONCLUSIONS:

The authors suggest that c-Met overexpression is associated with shorter survival time and poor treatment response in glioblastomas, the mechanism for which is elevated tumor invasiveness on the molecular and clinical phenotypes. This implies that more effective therapeutic strategies targeting c-Met receptors may have important clinical implication. Cancer 2009. © 2008 American Cancer Society.

Glioblastomas (GBMs) are a very aggressive tumor, and the prognosis for patients with GBMs remains poor despite multidisciplinary treatment approaches. Recently, the molecular and genetic characteristics underlying the pathogenesis and treatment resistance of these tumors have been under active investigation. Identifying and ultimately targeting the mechanisms by which gliomas resist cytotoxic agents will likely have a substantial impact on future treatment strategies. c-Met, in particular, is a transmembrane receptor protein tyrosine kinase (RPTK) for which the primary stimulatory ligand is hepatocyte growth factor (HGF), which is also known as scatter factor (SF). The c-Met RPTK has been implicated in the development and progression of several human cancers, such as hepatocellular carcinoma, osteosarcoma, colorectal cancer, and GBMs.1–9 Furthermore, c-Met is expressed in a wide variety of human carcinomas, in which its overexpression or misexpression often correlates with poor prognosis.1 In these tumors, c-Met signals have been known to induce proliferation, motility, cell adhesion, invasion, and antiapoptotic responses.2–6 c-Met affects tumorigenicity and malignant progression by inducing cell cycle progression, cell migration, and tumor angiogenesis.4, 5, 7–10 c-Met expression also inhibits apoptosis of tumor cells and confers resistance to cell death by chemoirradiation.4, 8, 11 Because of this clinical relevance and its multifunctional autocrine or paracrine tumor-promoting activities, SF/HGF/c-Met protects against death induced by a variety of DNA-damaging agents and ionizing radiation both in vitro and in vivo.4, 12–14 To our knowledge to date, the role of c-Met signaling in mediating radiation resistance in GBMs has been suggested in preclinical models,14 but has yet to be studied clinically. With regard to these aspects, c-Met-targeted treatment may be a promising therapeutic strategy.

Herein, we correlated c-Met expression in tumor specimens with overall survival and progression-free survival in patients with GBMs. Furthermore, to explain the tumor progression mediated by c-Met expression, we analyzed the radiologic and biologic associations between invasive phenotypes and c-Met overexpression.

MATERIALS AND METHODS

Patient and Tissue Collection

Sixty-six patients with GBMs were included in the current study. All patients underwent surgery at Samsung Medical Center between January 2004 and December 2006. Patients were managed according to established diagnostic and therapeutic protocols, including surgical resection and subsequent chemoradiotherapy. For the evaluation of histologic specimens, a macroscopic total resection was performed in 47 of 62 patients (75.8%), a partial resection in 14 of 62 patients (22.6%), and a biopsy only in 1 of 62 patients (1.6%). All patients in the intention-to-treated group underwent subsequent radiotherapy (60 grays [Gy] in 2-Gy fractions) after surgical resection. For concurrent chemoradiotherapy or adjuvant chemotherapy, 50 of 62 patients received temozolomide with a median of 4 cycles (range, 1-9 cycles) adminstered. However, 12 others (20%) did not receive chemotherapy because of clinical deterioration during radiotherapy. The response criteria described by Couldwell et al were used with modification.15 Tumor samples were reevaluated by neuropathologists (S.-Y.S. and Y.-L.S.) to confirm the diagnosis according to World Health Organization criteria. Tumor samples were obtained during surgical treatment and were embedded in paraffin for histologic studies. Written informed consent was obtained from all patients, and tissue collection was approved by each institutional review board. The patients' characteristics are shown in Table 1. Clinical data were obtained from the Samsung Medical Center tumor registry and from medical charts. These data included patient age at diagnosis, treatment type, treatment dates, and clinical and survival outcomes. Immunohistochemical (IHC) analysis was performed in a blinded manner, without prior knowledge of clinical outcome.

Table 1. Patient Characteristics
CharacteristicNo.
  • 95% CI indicates 95% confidence interval; MGMT, O-6-methylguanine-DNA methyltransferase.

  • *

    The number of available tissue samples was limited.

Patients62
Sex 
 Male36
 Female26
Mean age (range), y56 (13-78)
Mean follow-up period (range), mo13.3 (4.0-30.4)
Median overall survival (95% CI), mo14.3 (10.3-18.3)
Median progression-free survival (95% CI), mo6.3 (4.8-7.9)
Radiographic pattern 
 Solitary lesion46
 Invasive and multifocal lesions16
Surgical resection 
 Macroscopic total resection47
 Partial resection14
 Biopsy only1
Chemotherapeutic regimens 
 Temozolomide50
 No treatment12
MGMT status* 
 Unmethylated23
 Methylated21

IHC Study

Four-micrometer-thick sections sliced from paraffin-embedded specimens were prepared on the slide. Sections were immunostained with mouse monoclonal antibodies to c-Met (clone: 3D4; Zymed, Invitrogen, Carlsbad, Calif), c-Myc (clone 9E10.3; Neomarker, Thermo Fisher Scientific, Waltham, Mass), c-K-Ras (F234, sc-30; Santa Cruz Biotechnology, Santa Cruz, Calif), and c-Jun (Santa Cruz Biotechnology) and with rabbit polyclonal antibodies to matrix metalloproteinase (MMP)-2 (#AB807; Chemicon, Millipore, Billerica, Mass) and MMP-9 (#AB13458; Chemicon, Millipore). Tumor-containing sections were baked at 60°C for 30 minutes, deparaffinized in xylene, and rehydrated in graded concentrations of ethanol. Endogenous peroxidase activity was blocked by incubation in 0.3% hydrogen peroxide in methanol and with heat-induced antigen retrieval (for p53, 10 mM citrate buffer [pH 6.0] for 25 minutes in a vegetable steamer). Immunostaining involved sequential applications of primary antibody (c-Met at 1:50, MMP-2 at 1:200, MMP-9 at 1:100, c-Myc at 1:50, c-Jun at 1:40, and c-K-Ras at 1:10) for 16 hours at 4°C, followed by biotinylated secondary antibodies (Vector Laboratories, Orton Southgate, UK) at a 1:100 dilution for 1 hour and avidin-biotin complex (Elite ABC; Vector Laboratories) for 1 hour. Negative control slides received normal mouse serum (Dako Corporation, Carpinteria, Calif) as the primary antibody. Diaminobenzidine tetrahydrochloride was used as the enzyme substrate to observe the specific antibody localization, and Harris hematoxylin was used as a nuclear counterstain.

Scoring and Interpretation of Immunohistochemistry

Sections were examined for immunoreactivity for c-Met and other associated proteins, including c-K-Ras, c-Myc, c-Jun, MMP-2, and MMP-9, by an observer who was unaware of the histologic diagnoses, outcomes, or clinical features. Tumors were categorized as expressing little or no protein (a grade of 0 or 1), expressing levels similar to those in normal brain, or as overexpressing, with staining observed in a sizable subgroup of cells (25% to 50%; grade 2), most cells (50% to 75%; grade 3), or nearly all cells (>75%; grade 4) in the high-power field in areas with maximal staining. Tumors demonstrating moderate or strong c-Met immunopositivity in >25% of tumor cells were considered to be positive tumors. The expression of c-Met, c-K-Ras, c-Myc, c-Jun, MMP-2, and MMP-9 was analyzed as a dichotomous covariate: no or little immunoreactivity (a grade of either 0 or 1) versus overexpression (grades 2-4).

Bisulfite Modification and Methylation-specific Polymerase Chain Reaction

DNA from both the primary tumor and the cell lines was subjected to bisulfite treatment. Briefly, 1 μg of DNA was denatured by sodium hydroxide and modified by sodium bisulfite. DNA samples were then purified using the Wizard purification resin (Promega, Madison, Wis), treated again with sodium hydroxide, precipitated with ethanol, and resuspended in water. DNA methylation patterns in the CpG islands of O-6-methylguanine-DNA methyltransferase (MGMT) gene were determined by chemical treatment with sodium bisulfite and subsequent methylation-specific PCR (MSP). Primer pairs specific for methylated:5′tttcgacgttcgtt cgtaggttttcgc3′(sense), 5′gcactcttcc gaaaacgaaacg3′(anti- sense) and unmethylated: 5′tttgtg ttttgatgtttgtaggttttt gt3′ (sense), 5′aactccacactcttccaaa aacaaaaca3′ (antisense) were prepared. Polymerase chain reaction (PCR) conditions were 95°C for 5 minutes, and 95°C for 40 seconds, 60°C for 40 seconds, and 72°C for 40 seconds for 35 cycles, and 72°C for 5 minutes. Each PCR product was directly loaded on 8% acrylamide gel, stained with ethidium bromide, and observed under ultraviolet illumination.

Statistical Analysis

The analysis of the relation between c-Met expression and overall survival was calculated by using the Kaplan-Meier method (SPSS statistical software, version 11.0; SPSS Inc., Chicago, Ill). The differences between the survival curves were tested by using the log-rank test. The multivariate survival analysis was calculated according to the Cox proportional hazards model in a forward stepwise manner. The results are reported as being statistically significant if a 2-sided P value was <.05. The Fisher exact test was used to analyze the associations between dichotomous pathologic variables (ie, high MMP-2 and MMP-9 status, oncogenes status).

RESULTS

Clinical Outcome

The median age of the patients in the cohort at the time of diagnosis was 56 years (range, 13 years-78 years). There were 36 men and 26 women. The mean follow-up period at the time of analysis was 13.3 months (range, 4.0 months-30.4 months), and the median overall survival was 14.3 months (95% confidence interval [95% CI], 10.3 months-18.3 months) (Fig. 1A). At the end of follow-up, 13 (21.0%) patients remained free of disease progression. The median progression-free survival was 6.3 months (95% CI, 4.8 months-7.9 months).

Figure 1.

Graphs show (A) overall survival (OS) of 62 patients with glioblastomas. (B) Survival curve according to c-Met expression. (C) Progression-free survival (PFS) curve plotted according to c-Met expression, and OS of patients according to (D) the extent of resection, (E) invasive lesions, and (F) age.

Expression of c-Met: Prognostic Implications

c-Met overexpression was detected in 21 of 62 (33.9 %) GBMs. A total of 41 tumors demonstrated grade 0 or 1 (no or little) immunoreactivity, 10 tumors were grade 2, 5 tumors were grade 3, and 6 tumors were grade 4. In patients whose tumors overexpressed c-Met (ie, grades 2-4), the median overall survival was 10.7 months (95% CI, 9.3 months-12.0 months), compared with a median survival of 20.0 months (95% CI, 13.5 months-26.5 months) in patients whose tumors had no or little expression of c-Met (log-rank test, P < .001) (Fig. 1B). Therefore, overexpression of c-Met was found to be significantly associated with shorter overall survival from the time of initial diagnosis. With regard to treatment response, progression-free survival was 3.2 months (95% CI, 1.5 months-4.9 months) in patients whose tumors overexpressed c-Met and 9.2 months (95% CI 3.9 months-14.5 months) in patients whose tumors demonstrated no or little expression of c-Met (log-rank test, P = .002) (Fig. 1C). The methylation-specific PCR identified 21 of 44 (47.7%) GBMs in which methylated MGMT promoter was present and 23 of 44 (52.3%) tumors in which only unmethylated MGMT promoter was detectable (Table 1). On univariate analysis, the extent of surgical resection, type and amount of chemotherapy, and MGMT status were not found to be correlated with patient outcome (Table 2) (Figs. 1D-F). On multivariate analysis, age >70 years and overexpression of c-Met were found to be the most significant independent predictors of poor overall survival (P = .019 and P = .015, respectively) (Table 3).

Table 2. Univariate Analysis of Overall Survival
Prognostic FactorsMedian OS, moP*
  • OS indicates overall survival; KPS, Karnofsky performance status; MGMT, O-6-methylguanine-DNA methyltransferase; MMP, matrix metalloproteinase.

  • *

    Log-rank test.

Age, y .019
 ≤7017.1 
 >709.8 
Sex .900
 Male13.7 
 Female15.8 
KPS .821
 >7016.2 
 ≤7012.9 
Extent of resection .072
 Macroscopic total resection18.2 
 Partial resection10.7 
Radiographic pattern .004
 Solitary lesion18.2 
 Invasive and multifocal lesions10.7 
Chemotherapy .687
 Temozolomide (concurrent or adjuvant)16.7 
 No treatment11.2 
Status of MGMT .313
 Unmethylated13.7 
 Methylated20.0 
Expression of c-Met <.001
 No or little expression20.0 
 Overexpression10.7 
Expression of c-Myc .300
 No or little expression13.3 
 Overexpression20.0 
Expression of MMP-2 .183
 No or little expression27.0 
 Overexpression13.7 
Expression of MMP-9 .984
 No or little expression17.1 
 Overexpression13.3 
Expression of c-K-Ras .245
 No expression18.2 
 Overexpression13.7 
Expression of c-Jun .388
 No or little expression17.1 
 Overexpression13.2 
Table 3. Multivariate Analysis of Overall Survival
Prognostic FactorsHazards Ratio (95% CI)P*
  • 95% CI indicates 95% confidence interval.

  • *

    Determined using the Cox proportional hazards regression model.

Aged >70 y2.748 (1.181-6.391).019
Expression of c-Met2.761 (1.214-6.280).015
Radiographic pattern2.278 (0.871-5.957).093
Extent of resection1.004 (0.389-2.595).993

Invasive and Multifocal Phenotypes

To clarify the invasive behaviors of tumors mediated by c-Met, we analyzed the association between c-Met expression and radiographic patterns. Magnetic resonance imaging demonstrated that 16 of 62 GBMs (29%) contained lesions with satellite-enhancing lesions (n = 12), accompanying subependymal/leptomeningeal seeding (n = 2), and multiple tumor involvements (n = 2). Those tumors were clinically relevant as having invasive and multifocal features, compared with 46 (74.2%) other tumors having solitary contrast-enhancing lesions. These invasive and multifocal phenotypes are known to be correlated with poor clinical outcome.15–19 For patients with tumors with invasive and multifocal phenotypes, the median survival was only 10.7 months (95% CI, 9.0 months-12.3 months), compared with a median survival of 18.2 months (95% CI, 11.0 months-25.4 months) for patients with solitary enhancing tumors (log-rank test, P = .004). In the analysis of association with c-Met expression, 9 of 21 tumors (42.9%) that overexpressed c-Met demonstrated invasive and multifocal features on the initial magnetic resonance images, whereas only 7 of 41 tumors (17.1%) having no or little c-Met expression demonstrated these features (Fisher exact test, P = .036).

Association of Expression of c-Met With Molecular Phenotypes

To explore the possible association between c-Met expression and tumor invasiveness, we performed MMP-2 and MMP-9 IHC analysis. MMP-2 and MMP-9 were highly expressed in 35 of 62 (56.5%) and 38 of 62 (61.3%) GBMs, respectively. Strong MMP-2 expression was observed in 16 of 21 (76.2%) c-Met-overexpressing tumors and in 19 of 41 (46.3%) tumors with no or little c-Met expression (Fisher exact test, P = .032) (Fig. 2). Similarly, high MMP-9 expression was detected in 17 of 21 (81.0%) c-Met-overexpressing tumors, compared with 21 of 41 (51.2%) tumors with no or little c-Met expression (Fisher exact test, P = .029). Therefore, c-Met expression was found to be significantly associated with molecular phenotypes representing invasiveness, such as the expression of MMP-2 and MMP-9 (Table 4).

Figure 2.

c-Met, matrix metalloproteinase (MMP)-2, immunohistochemical analysis, and radiographic pattern are shown. Representative images show (A) diffuse c-Met immunopositivity with (B) abundant MMP-2 immunoreactivity and (C) a radiographic pattern demonstrating multifocal lesions in 1 glioblastoma specimen (No. 218) and (D) a lack of c-Met immunopositivity with (E) weak MMP-2 immunoreactivity and its radiographic pattern demonstrating (F) solitary, contrast-enhancing lesions in another glioblastoma specimen (No. 119).

Table 4. Association Between c-Met Expression and Molecular Phenotypes
SampleNo c-Met ExpressionOverexpression of c-MetP*
  • MMP indicates matrix metalloproteinase.

  • *

    Determined using the Fisher exact test.

MMP-2  .032
 Positive1916 
 Negative225 
MMP-9  .029
 Positive2117 
 Negative204 
c-Myc  .002
 Positive1315 
 Negative286 
c-K-Ras  .086
 Positive3116 
 Negative105 
c-Jun  .053
 Positive59 
 Negative3612 

Oncogenes, such as activated c-Myc, c-Jun, and c-K-Ras, can induce c-Met overexpression through transcriptional mechanisms.20–24 To determine the oncogenic potential of c-Met, we next assessed c-Myc, c-Jun, and c-K-Ras expression by immunohistochemistry. Overexpression of c-Myc was observed in 15 of 21 (71.4%) c-Met-overexpressing tumors and in 13 of 41 (31.7%) tumors with no or little c-Met expression (Fisher exact test, P = .002). c-Myc expression was found to be significantly correlated with c-Met expression, whereas a borderline statistical correlation was found for c-Jun overexpression (P = .053) and c-K-Ras overexpression (P = .086) (Table 4) (Fig. 3).

Figure 3.

C-met and c-Myc immunohistochemical analysis is shown. Representative images show (A) diffuse c-Met immunopositivity and (B) abundant c-Myc immunoreactivity in 1 glioblastoma specimen (No. 309) and (C) a lack of c-Met immunopositivity with (D) a lack of c-Myc staining in another glioblastoma specimen (No. 064).

DISCUSSION

GBMs aggressively invade the surrounding brain, making curative surgical resection nearly impossible. To our knowledge, the current study is the first investigation aimed at evaluating the prognostic significance of c-Met expression in a series of patients with newly diagnosed GBMs. The HGF/SF/c-Met pathway is known to influence cellular proliferation via the PI3K/Akt pathway, increase invasiveness, and have antiapoptotic effects through several mechanisms.1, 8, 25, 26 Our results demonstrated that c-Met overexpression in tumors indicated a poor prognosis for patients with glioblastoma (10.7 months vs 20.0 months; P < .001). In the majority of cases, patients with initially diagnosed GBM underwent subsequent radiotherapy or concurrent chemoradiotherapy during the first 3 months after the initial diagnosis. Considering this therapeutic schedule, it was noteworthy that higher expression of c-Met was closely associated with shorter progression-free survival (3.2 months vs 9.2 months; P = .002) as well as with shorter overall survival. This significant influence on disease progression suggests that c-Met expression conferred resistance to the chemoirradiation in GBMs. This result is consistent with previous reports regarding radioresistance enhanced by c-Met expression.14 Therefore, to overcome the radioresistance induced by c-Met expression, it will be necessary to develop more effective therapeutic strategies targeting the c-Met receptor. Our results suggest that approximately 30% of GBM patients may be good candidates for c-Met-targeted treatment.

MMP-2 and MMP-9 have been shown to contribute to invasion and metastasis in diverse malignant neoplasms.27–31 To determine the role of the c-Met signal pathway in this mechanism, we analyzed the association between c-Met expression and MMP expression in 62 GBMs and found that the expression of MMP-2 and MMP-9 was elevated in approximately 60% of the cases. The immunoreactive proteins of MMP-2 and MMP-9 were localized to tumor cells and vasculature cells of tumors. Inflammatory cells (macrophages) also stained positively. In this study, MMP expression was significantly correlated with c-Met overexpression. Our findings agree with experimental observations of a role for MMP-9 and MMP-2 in glioma invasion in vitro and in animal invasion models.29, 32–35 Given that the c-Met signaling pathway has multifunctional mechanisms, our results suggest that c-Met may enhance MMP expression and activation in cancer cells.

However, we do acknowledge that our study had drawbacks, because MMP expression could not be accurately measured by immunohistochemistry. Our results represent only trends indicating increases or decreases in MMP expression. Accordingly, in an attempt to validate the potential role of c-Met signaling in tumor invasiveness, we analyzed radiographic patterns. Tumor invasiveness can be characterized by atypical radiographic patterns, including multifocal lesions and intracranial tumor disseminations.15–19 To our knowledge to date, little is known regarding the relation between c-Met expression and radiographic patterns. In the current study, 9 of 21 (42.9%) tumors with c-Met overexpression radiographically reflected invasive and multifocal features on the initial magnetic resonance images, compared with only 7 of 41 (17.1%) tumors with no or little c-Met expression. Therefore, this finding provides additional evidence of the mechanism by which c-Met affects survival and cancer progression.

Recent studies have demonstrated that tumor suppressor and related genes collaborate with c-Met expression, eventually resulting in carcinogenesis.5, 21 Our results suggesting an association between c-Met signaling and c-Myc expression are consistent with the work of those authors who reported that a significant proportion of hepatocellular carcinomas and GBMs coexpress c-Met and c-Myc. However, further study will be needed to validate the relation of oncogenes.

In conclusion, the results of the current study demonstrate that c-Met overexpression is associated with shorter survival time and poor treatment response in patients with GBMs. We also suggest that the mechanism by which c-Met expression affects prognosis may be associated with tumor invasiveness. c-Met expression in tumors can be assessed in routinely processed patient samples. Therefore, if we develop standards that can be applied across diagnostic laboratories, more effective therapeutic strategies targeting the c-Met receptor may have important clinical implications.

Acknowledgements

We gratefully acknowledge Mi Hyun Kim and Shi Yeon Kim for assistance with immunohistochemistry. We also thank Dr. Hyung-gi Kim for helpful discussions and critical reading of the article.

Ancillary